The following is a summary of “Depression prevalence of the Geriatric Depression Scale-15 was compared to Structured Clinical Interview for DSM using individual participant data meta-analysis,” published in the July 2024 issue of Psychiatry by Parsons et al.
The Geriatric Depression Scale (GDS-15), calibrated for diagnostic accuracy, is commonly used with a cutoff of ≥5 to estimate depression prevalence in older adults.
Researchers conducted a retrospective study evaluating the accuracy of the GDS-15 with a cutoff of ≥5 in estimating depression prevalence compared to Structured Clinical Interview for Diagnostic and Statistical Manual (SCID) diagnoses and to identify if an alternative cutoff might offer greater precision.
They integrated data from 14 studies with 3,602 participants, of whom 434 were diagnosed with major depression by SCID. Generalized linear mixed models were used to estimate depression prevalence according to both GDS-15 ≥5 and SCID diagnoses.
The result showed that the pooled prevalence of depression using the GDS-15 ≥5 cutoff was 34.2% (95% CI 27.5–41.6%), whereas the SCID prevalence was 14.8% (95% CI 10.0–21.5%), reflecting a difference of 17.6% (95% CI 11.6–23.6%). A GDS-15 cutoff of ≥8 yielded a prevalence estimate closest to SCID, with a mean difference of −0.3% (95% prediction interval −17.0–16.5%).
Investigators concluded that GDS-15 with a cutoff of ≥5 overestimated depression prevalence, whereas ≥8 was more accurate, though heterogeneity limits practical use, with validated diagnostic interviews remaining the gold standard.
Source: nature.com/articles/s41598-024-68496-3
